Neurons of the medullary rostral ventral respiratory group (rVRG) represent the primary premotor monosynaptic, excitatory input to the C3- C5 phrenic motoneuron pool. Hemisection of the spinal cord at C2 interrupts this descending bulbospinal projection thereby resulting in an irreversible, ipsilateral hemiparesis of the diaphragm. If animals are presented with a respiratory challenge (e.e., hypoxia), then function returns to the paralyzed hemidiaphragm by activation of previously functionally-silent rVRG axons that cross at the level of the phrenic motoneuron pool (i.e., the crossed phrenic phenomenon, (CPP). This combination of a physiologically direct and anatomically well- characterized descending inputs to a motoneuron pool having a high degree of synaptic integration and an inherent potential for neuroplastic responses offers an important experimental and clinically relevant setting in which to explore (i) the capacity of a cellular intervention to restore supraspinal activation of motoneurons below the level of injury and (ii) possible positive or adverse effects of cellular replacement on spontaneous repair properties. Thus, Aim 1 will test whether respiratory-associated phrenic motoneuronal discharges are restored following the introduction of fetal tissue into hemiresection cavities made at the C2 spinal level and whether expression of the CPP is affected by this type of graft. Emphasis will be placed on testing the contribution of these grafts as functional relays between separated regions of the spinal cord using neuroanatomical and electrophysiological methods. From a clinical perspective, however, contusive trauma to the spinal cord is a more frequent occurrence, and this can result in some sparing of white and/or gray matter depending on the severity and spinal level of injury. In either case, some residual, though weakened, function would be expected. Aim 2 will determine to what extent fetal grafts can effectively enhance such spared respiratory function in chronic contusive injuries at C2 and acute and chronic contusion lesions at C4-C5. The latter injuries also will afford an opportunity to investigate injury- and graft-associated changes in the activities of phrenic and abdominal muscle motoneurons that are involved in the cough reflex which is often impaired by cervical spinal cord injury. Lastly, immunological deletion of grafts will be used in Aim 3 to directly assess fetal graft contributions to altered motoneuronal activity in the resection cervical spinal cord injury model. Whether graft rejection precipitates an augmented loss of phrenic motoneuron function also will be examined.